Content uploaded by Naranbaatar Dashdorj
Author content
All content in this area was uploaded by Naranbaatar Dashdorj on Sep 01, 2018
Content may be subject to copyright.
IN THIS ISSUE OF
HEPATOLOGY
Volume 66 !Number 6 December 2017
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
Hepatology Highlights
1709 Clara Tow, Robert S. Brown Jr.,
Zaid Husni Tafesh, Robert E. Schwartz,
Russell Rosenblatt, Shawn L. Shah,
Nicole T. Shen, Aleksey Novikov,
Shirley Cohen-Mekelburg, Nicholas Russo,
and Vikas Gupta
Editorials
1713 Fecal Microbiota Transplantation for Hepatic
Encephalopathy: Ready for Prime Time?
Puneeta Tandon, Karen Madsen, and Dina Kao
(See Article on Page 1727)
1716 Delta Hepatitis: Toward Improved Diagnostics
Saleem Kamili, Jan Drobeniuc,
Tonya Mixson-Hayden, and Maja Kodani
(See Article on Page 1739)
1719 Polo-Like-Kinase 1: A Key Cellular Target for
Anti-HBV Therapy?
Michael M.C. Lai and Wen-Chi Su
(See Article on Page 1750)
1722 Redefining Successful Treatment of Severe
Alcoholic Hepatitis
Michael Ronan Lucey
(See Article on Page 1842)
1724 Beyond Farnesoid X Receptor to Target New
Therapies for NAFLD
Xiaoying Liu and Richard M. Green
(See Article on Page 1854)
Rapid Communication
1727 *Fecal Microbiota Transplant From a
Rational Stool Donor Improves Hepatic
Encephalopathy: A Randomized
Clinical Trial
Jasmohan S. Bajaj, Zain Kassam, Andrew Fagan,
Edith A. Gavis, Eric Liu, I. Jane Cox,
Raffi Kheradman, Douglas Heuman,
Jessica Wang, Thomas Gurry, Roger Williams,
Masoumeh Sikaroodi, Michael Fuchs, Eric Alm,
Binu John, Leroy R. Thacker, Antonio Riva,
Mark Smith, Simon D. Taylor-Robinson,
and Patrick M Gillevet
(See Editorial on Page 1713)
Original Articles
VIRAL HEPATITIS
1739 *A Novel Quantitative Microarray Antibody
Capture Assay Identifies an Extremely High
Hepatitis Delta Virus Prevalence Among
Hepatitis B Virus–Infected Mongolians
Xiaohua Chen, Odgerel Oidovsambuu, Ping Liu,
Rosslyn Grosely, Menashe Elazar,
Virginia D. Winn, Benjamin Fram, Zhang Boa,
Hongjie Dai, Bekhbold Dashtseren,
Dahgwahdorj Yagaanbuyant, Zulkhuu Genden,
Naranbaatar Dashdorj, Andreas Bungert,
Naranjargal Dashdorj, and Jeffrey S. Glenn
(See Editorial on Page 1716)
*Human Study
>Cover Figure: Novel
Quantitative Micro Array
Capture (Q-MAC) assay reveals
extremely high prevalence rate of
hepatitis delta virus (HDV) in
Mongolia, contributing to the
world’s highest hepatocellular
carcinoma (HCC) rate. See
article on page 1739.
Delta Hepatitis: Toward Improved
Diagnostics
SEE ARTICLE ON PAGE 1739
Hepatitis D virus (HDV), the etiologic agent
of hepatitis D or delta hepatitis, is a unique
human virus that requires the hepatitis B
surface antigen (HBsAg) of hepatitis B virus (HBV)
for its replication and to establish infection. HDV, the
only member of its own separate genus, Deltavirus, is
the smallest known infectious viral agent of humans.
(1)
The HDV virion, which measures 35-37 nm in diame-
ter, contains a single-stranded circular !1,700-nucleo-
tide RNA genome of negative polarity, which encloses
only a single functional open reading frame encoding
the sole viral protein, the hepatitis delta antigen
(HDAg). The HDAg is associated with the genomic
RNA to form a ribonucleoprotein complex which is
encapsulated by a lipid envelope containing the small,
middle, and large HBsAg proteins.
(2)
HDV exhibits a
high degree of genetic heterogeneity, with estimated
mutation rates between 3 310
–2
and 3 310
–3
base
substitutions per genomic site per year.
(1)
The virus has
a wide geographic distribution with eight distinct
genotypes. HDV genotype 1 is the most common
genotype, being prevalent in North America, Europe,
the Middle East, and North Africa; genotypes 2 and
4 are found predominantly in East Asia; genotype 3 is
found exclusively in the Amazon basin in South Amer-
ica; and genotypes 5-8, also known as African geno-
types, are predominantly found in West and central
Africa.
(3)
HDV is a highly pathogenic virus, and clinical pre-
sentation of hepatitis D ranges from fulminant hepati-
tis, exacerbation of the course of underlying HBV
infection, and acceleration of progression to cirrhosis,
leading to early decompensation of liver function and
hepatocellular carcinoma in the majority of patients.
(4)
However, a benign course of HDV infection has also
been observed in Greece, Samoa, and the Far East;
whether this is related to various viral characteristics
such as the infecting genotype or host genetics remains
to be determined.
(4)
Outbreaks of severe and fulminant
hepatitis D have been reported from Brazil, Russia,
Greenland, and Mongolia.
(3)
The laboratory diagnosis
of coinfection or superinfection with HDV is based on
simultaneous detection of various serologic markers of
HBV and HDV infection. The markers of acute HDV
infection include, along with the markers of HBV
infection, HDAg, HDV RNA, and immunoglobulins
M and G (IgM and IgG, respectively) antibodies to
HDV (anti-HDV). These markers of HDV infection
are present only transiently and disappear during early
convalescence; IgM anti-HDV and even IgG anti-
HDV also disappear with time in acute resolving
HDV infection.
(3)
Superinfection of HBsAg carriers,
which almost always leads to chronic hepatitis D, is
marked by absence of IgM antibody to hepatitis B core
antigen and presence of all the other markers of HBV
and HDV infections (Fig. 1). However, markers of
HBV replication, especially HBV DNA, may be sup-
pressed during the acute phase of HDV infection and
remain undetectable.
(1)
Chronicity of HDV infection
is associated with persistence of HDAg, HDV RNA,
and IgM and IgG anti-HDV. HDV RNA is the gold
standard for diagnosis of current HDV infection
because assays for detection of HDAg are fraught with
sensitivity and specificity issues. Quantitative assays of
HDV RNA are useful for monitoring response to anti-
viral therapy, and the recent availability of a World
Abbreviations: anti-HDV, antibody to HDV; HBsAg, hepatitis B
surface antigen; HBV, hepatitis B virus; HDAg, hepatitis delta anti-
gen; HDV, hepatitis D virus; IgG/IgM, immunoglobulins G/M;
Q-MAC, quantitative microarray antibody capture.
Received July 31, 2017; accepted September 27, 2017.
Published 2017. This article is a U.S. Government work and is in
the public domain in the USA.
View this article online at wileyonlinelibrary.com.
DOI 10.1002/hep.29564
Potential conflict of interest: Nothing to report.
ADDRESS CORRESPONDENCE AND
REPRINT REQUESTS TO:
Saleem Kamili, Ph.D.
Division of Viral Hepatitis, Centers for Disease Control
and Prevention
1600 Clifton Road
Atlanta, GA 30329-4027
E-mail: sek6@cdc.gov
1716
HEPATOLOGY, VOL. 66, NO. 6, 2017
A
HE STU
DYOFLIVERD I S
E ASES
T
ME
RIC
A
N
A
SSOCI
ATIO
NFOR
Health Organization standard has further helped in
optimizing assays across various centers and laborato-
ries. Sequencing of HDV RNA–positive samples is a
reliable way to determine the HDV genotype.
(5)
The prevalence of HDV is a measure of anti-HDV
positivity among HBsAg positive carriers. It is esti-
mated that 10 million to 20 million individuals (!5%
of chronic HBV patients) worldwide are coinfected
with HDV.
(6)
Vaccination against hepatitis B, which is
also the default vaccination against HDV, resulted in a
decrease of HDV prevalence in industrialized countries
with the implementation of routine hepatitis B immu-
nization of children and other populations.
(7)
How-
ever, in several European countries (e.g., Germany,
Italy, and England), which have large and increasing
numbers of migrants from HDV-endemic areas (e.g.,
eastern Europe and Turkey), HDV prevalence rates
have remained unchanged.
(2)
High-risk populations
like injection drug users continue to get impacted by
superinfection with HDV, as has been observed in
Europe and recently in the United States.
(7,8)
Based on
cross-sectional studies, high rates of HDV infection,
ranging from 10% to 70%, in HBsAg carriers have
been reported from Nigeria, Gabon, India, Pakistan,
Iran, the western Brazilian Amazon, Tajikistan, and
Mongolia.
(6)
A number of developing countries have
reported a high endemicity of HDV, and prevalence
rates of >20% have been reported; these include cen-
tral Africa, mountainous regions of Venezuela and
Colombia, Romania, Pakistan, Iran, the Amazon basin
in South America, and Mongolia. However, reliable
data on the accurate prevalence of HDV are not avail-
able from all regions due to either lack of testing of
HBsAg carriers for HDV infection or lack of availabil-
ity of anti-HDV antibody assays with proven perfor-
mance characteristics. In this context, the article by
Chen et al. published in this issue of HEPATOLOGY
(9)
is
of major significance. The authors found a substantially
higher prevalence (!60%) of HDV infections among
HBV-infected individuals identified in a national survey
sampling of the Mongolian population. Of a total of
1,158 individuals chosen based on a three-stage cluster
sampling method to reflect the gender, age, and
geographical origin representative of the entire country,
123 tested positive for HBsAg, of whom 75 (60%)
tested anti-HDV-positive by a newly developed high-
throughput quantitative microarray antibody capture
(Q-MAC) assay that the authors describe.
(9)
Modern
diagnostic approaches, such as protein microarray–based
Q-MAC assays, allow for improvement in the sensitivi-
ties of various diagnostic assays. Protein microarrays
have become an important tool in studies of protein–
protein interactions, protein detection, and other appli-
cations in quantitative and functional proteomics and
beyond. Protein microarrays are arrays of protein targets,
frequently antibodies and/or antigens, immobilized on a
solid support such as a glass slide. Plasmonic substrates,
such as nanostructured plasmonic gold film, have been
"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
FIG. 1. Serologic course of acute resolving (A) and chronic (B) HDV infection. Abbreviations: anti-HBc, antibody to hepatitis B
core antigen; anti-HBs, antibody to HBsAg.
"""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""
HEPATOLOGY, Vol. 66, No. 6, 2017 KAMILI ET AL.
1717
developed to improve the sensitivity and dynamic range
of protein detection on microarrays and have been
shown to detect cancer biomarkers and integrated
human antibodies and antigens down to femtomolar
ranges.
(10)
Chen et al. have used this protein microarray
technology to develop a Q-MAC assay for sensitive
quantitative fluorescent detection of anti-HDV IgG
from patient sera. The authors not only convincingly
demonstrated the excellent performance characteristics
of their Q-MAC assay but also established quantitative
thresholds of captured HDV antibody predictive of
HDV RNA positivity. However, as the authors have
righty stated, despite a good correlation between fluo-
rescence intensity of the Q-MAC assay and HDV
RNA levels, determination of HDV RNA remains the
gold standard for monitoring of treatment against hepa-
titis D. The Q-MAC assay is rapid, requires a small
volume of patient serum, and can easily be scaled to
high-throughput screening of large cohorts of HBsAg
carriers for HDV infection.
A true seroepidemiologic gauge of the prevalence of
hepatitis D is to test HBsAg carriers only and estimate
the prevalence rates using the HBsAg positives as the
denominator. With this approach, Chen et al.
(9)
found
an alarming proportion of their HBsAg-positive popu-
lation in Mongolia superinfected with HDV. Com-
pared to other HBsAg carriers, persons superinfected
with HDV are at the highest risk for hepatocellular
carcinoma, and HDV superinfection undoubtedly con-
tributes to Mongolia’s high rate of hepatocellular carci-
noma. Assessments of HDV as a cause of excess
morbidity and mortality among persons with HBV
infection globally are limited by the lack of serologic
studies. The data from Chen et al.
(9)
highlight the
importance of conducting such seroprevalence studies
among HBsAg carriers in HDV-endemic countries
and high-risk populations in developed countries.
Given that an estimated 248 million are chronically
infected with HBV worldwide and thus susceptible to
superinfection with HDV, concerted efforts, like the
one undertaken by the Hepatitis Delta International
Network (http://hepatitis-delta.org/), are needed for
the determination of accurate estimates of hepatitis D
disease burden, evidence-based policies for HDV
testing, and research to understand the biological
mechanisms of HDV infection and find efficacious
therapies for treatment of hepatitis D.
Acknowledgment: We thank Dr. John Ward, Division
of Viral Hepatitis, Centers for Disease Control and
Prevention, for a critical review of the manuscript.
Saleem Kamili, Ph.D.
Jan Drobeniuc, M.D., Ph.D.
Tonya Mixson-Hayden, Ph.D.
Maja Kodani, Ph.D.
Division of Viral Hepatitis
Centers for Disease Control and Prevention
Atlanta, GA
REFERENCES
1) Taylor JM, Purcell RH, Farci P. Hepatitis D (delta) virus. In:
Knipe DW, Howley PM, eds. Fields Virology, Vol 2, 6th ed.
Philadelphia: Lippincott Williams & Wilkins; 2013:2222-2241.
2) Lempp FA, Ni Y, Urban S. Hepatitis delta virus: insights into a
peculiar pathogen and novel treatment options. Nat Rev Gastro-
enterol Hepatol 2016;13:580-589.
3) Rizzetto M. Hepatitis D virus: introduction and epidemiology.
Cold Spring Harb Perspect Med 2015;5:a021576.
4) Sureau C, Negro F. The hepatitis delta virus: replication and
pathogenesis. J Hepatol 2016;64:S102-S116.
5) Kodani M, Martin A, Mixson-Hayden T, Drobeniuc J, Gish
RR, Kamili S. One-step real-time PCR assay for detection and
quantitation of hepatitis D virus RNA. J Virol Methods 2013;
193:531-535.
6) Hughes SA, Wedemeyer H, Harrison PM. Hepatitis delta virus.
Lancet 2011;378:73-85.
7) Holmberg SD, Ward JW. Hepatitis delta: seek and ye shall find.
J Infect Dis 2010;202:822-824.
8) Ahn J, Gish RG. Hepatitis D virus: a call to screening. Gastro-
enterol Hepatol (NY) 2014;10:647-686.
9) Chen X, Oidovsambuu O, Liu P, Grosely R, Elazar M, Winn
VD, et al. A novel quantitative microarray antibody capture
(Q-MAC) assay identifies an extremely high HDV prevalence
amongst HBV infected Mongolians. HEPATOLOGY 2017;66:
1739-1749.
10) Tabakman SM, Lau L, Robinson JT, Price J, Sherlock SP,
Wang H, et al. Plasmonic substrates for multiplexed protein
microarrays with femtomolar sensitivity and broad dynamic
range. Nat Commun 2011;2:466.
KAMILI ET AL. HEPATOLOGY, December 2017
1718
A Novel Quantitative Microarray
Antibody Capture Assay Identifies an
Extremely High Hepatitis Delta Virus
Prevalence Among Hepatitis B
Virus–Infected Mongolians
Xiaohua Chen,
1*
Odgerel Oidovsambuu,
2,3*
Ping Liu,
1*
Rosslyn Grosely,
1
Menashe Elazar,
1
Virginia D. Winn,
4
Benjamin Fram,
1
Zhang Boa,
5
Hongjie Dai,
5
Bekhbold Dashtseren,
2,6,7
Dahgwahdorj Yagaanbuyant,
2,6,7
Zulkhuu Genden,
2,6
Naranbaatar Dashdorj,
6
Andreas Bungert,
6
Naranjargal Dashdorj,
2,6
and Jeffrey S. Glenn
1,8,9
Hepatitis delta virus (HDV) causes the most severe form of human viral hepatitis. HDV requires a hepatitis B virus (HBV)
coinfection to provide HDV with HBV surface antigen envelope proteins. The net effect of HDV is to make the underlying
HBV disease worse, including higher rates of hepatocellular carcinoma. Accurate assessments of current HDV prevalence
have been hampered by the lack of readily available and reliable quantitative assays, combined with the absence of a Food and
Drug Administration–approved therapy. We sought to develop a convenient assay for accurately screening populations and to
use this assay to determine HDV prevalence in a population with abnormally high rates of hepatocellular carcinoma. We
developed a high-throughput quantitative microarray antibody capture assay for anti-HDV immunoglobulin G wherein
recombinant HDV delta antigen is printed by microarray on slides coated with a noncontinuous, nanostructured plasmonic
gold film, enabling quantitative fluorescent detection of anti-HDV antibody in small aliquots of patient serum. This assay was
then used to screen all HBV-infected patients identified in a large randomly selected cohort designed to represent the Mongo-
lian population. We identified two quantitative thresholds of captured antibody that were 100% predictive of the sample either
being positive on standard western blot or harboring HDV RNA detectable by real-time quantitative PCR. Subsequent
screening of the HBV
1
cohort revealed that a remarkable 57% were RNA
1
and an additional 4% were positive on western
blot alone. Conclusion: The quantitative microarray antibody capture assay’s unique performance characteristics make it ideal
for population screening; its application to the Mongolian HBV surface antigen–positive population reveals an apparent
!60% prevalence of HDV coinfection among these HBV-infected Mongolian subjects, which may help explain the extraordi-
narily high rate of hepatocellular carcinoma in Mongolia. (HEPATOLOGY 2017;66:1739-1749).
SEE EDITORIAL ON PAGE 1716
Hepatitis delta virus (HDV) causes the most
rapidly progressive human viral hepatitis,
leading to accelerated rates of cirrhosis and
hepatocellular carcinoma.
(1,2)
HDV has a unique 1.7-
kB single-stranded circular RNA genome that encodes
for one protein, the viral coat–like protein hepatitis
delta antigen (HDAg). Together, these are encapsi-
dated by a lipid envelope containing hepatitis B virus
(HBV) surface antigen (HBsAg) envelope proteins
Abbreviations: anti-HDV, antibody to HDV; ELISA, enzyme-linked immunosorbent assay; FBS, fetal bovine serum; HBsAg, HBV surface antigen;
HBV, hepatitis B virus; HCV, hepatitis C virus; HDAg, hepatitis delta antigen; HDV, HDV, hepatitis delta virus; IgG, immunoglobulin G;
pGOLD, plasmonic gold; Q-MAC, quantitative microarray antibody capture; S-HDAg, small hepatitis delta antigen.
Received June 27, 2016; accepted November 22, 2016.
Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.28957/suppinfo.
X. Chen’s present address is: Department of Infectious Disease, Shanghai Jiao Tong University Affiliated Sixth People’s Hospital, Shanghai, China.
*These authors contributed equally to this work.
Supported by the National Institutes of Health (RUL1RR025780 and R01HD060723, to V.D.W.). X.C. was supported by a Visiting Research Scientist
award.
1739
HEPATOLOGY, VOL. 66, NO. 6, 2017 VIRAL HEPATITIS
A
HE STU
DYOFLIVERD I S
E ASES
T
ME
RIC
A
N
A
SSOCI
ATIO
NFOR
that HDV acquires from HBV and that replicates in
the hepatocytes simultaneously with HDV.
(3)
This
requirement for HBV envelope proteins is the only
helper function provided by HBV but explains why
HDV can only infect subjects with a coexisting HBV
infection due either to the simultaneous transmission
of the two viruses or to superinfection in an established
HBV carrier.
(4)
Approximately 5% of the global HBV-
infected population, or 15 million to 20 million people
worldwide, are infected with HDV, although HDV
prevalence rates are not uniform, with higher rates of
HDV coinfection reported in the Mediterranean basin,
parts of Africa, the Middle East, and South Ameri-
ca.
(5)
In a study of 249 apparently healthy individuals
living in and around the capital city of Mongolia, 10%
were HBsAg
1
, with 83% of those having detectable
HDV RNA,
(6)
prompting calls for a larger nationwide
survey.
The usual first step in the diagnosis of HDV infec-
tion is testing HBsAg
1
individuals for antibody to
HDV (anti-HDV). Anti-HDV is not protective; it is
present in all immunocompetent patients with HDV
infection.
(7)
Total antibodies to HDV can be detected
with an enzyme-linked immunosorbent assay
(ELISA). In anti-HDV
1
patients, the ideal next step
is testing for HDV RNA in serum to confirm the pres-
ence of active HDV infection. With the advent of RT-
PCR techniques, HDV RNA has been measured with
qualitative or semiquantitative RT-PCR assays.
(8)
Sen-
sitivity has markedly improved, with current detection
limits of 1,000 genome/mL for simple PCR and 10
genome/mL for nested PCR.
(8)
Unfortunately, the
results from different laboratories are often not compa-
rable due to the diverse sensitivity of the assays; vari-
ance is caused by the use of different primer sets and
by the variability of the RNA region amplified.
(9,10)
A
World Health Organization international RNA stan-
dard is now available, enabling the reporting of results
in international units, although no quantitative HDV
RNA assay is commercially available in the United
States.
Here, we present a new quantitative microarray anti-
body capture (Q-MAC) assay for detecting anti-HDV
in human sera. This platform is constructed on noncon-
tinuous, nanostructured plasmonic gold (pGOLD)
films with enhanced near-infrared fluorescence detec-
tion that we hypothesized would have high sensitivity
and would be ideal for high-throughput antibody cap-
ture screening. Indeed, similar technology was previous-
ly demonstrated to have vastly improved sensitivity over
peptide arrays on glass, with the limit of detection down
to the 10 femtomolar (picograms per milliliter)
range.
(11,12)
For the anti-HDV Q-MAC assay, recom-
binant full-length HDV small delta antigen (S-HDAg)
was arrayed on a pGOLD substrate for sensitive profil-
ing of antibodies in the sera of HDV patients. We first
determined the performance characteristics of this new
assay format using reference HDV RNA–positive and
negative control sera. We then used the assay to deter-
mine the prevalence of HDV coinfection among HBV-
infected individuals identified in a national survey sam-
pling the population of Mongolia.
Copyright V
C2016 by the American Association for the Study of Liver Diseases.
View this article online at wileyonlinelibrary.com.
DOI 10.1002/hep.28957
Potential conflict of interest: Nothing to report.
ARTICLE INFORMATION:
From the
1
Department of Medicine, Division of Gastroenterology and Hepatology, Stanford University School of Medicine, Palo Alto,
CA;
2
Liver Center and
3
Mongolian National University, Ulaanbaatar, Mongolia;
4
Department of Obstetrics and Gynecology and
5
Department of Chemistry, Stanford University, Palo Alto, CA;
6
Onom Foundation and
7
Mongolian National University of Health
Sciences, Ulaanbaatar, Mongolia;
8
Department of Microbiology and Immunology, Stanford University School of Medicine, Palo Alto, CA;
9
Veterans Administration Medical Center, Palo Alto, CA.
ADDRESS CORRESPONDENCE AND REPRINT REQUESTS TO:
Jeffrey S. Glenn, M.D., Ph.D.
Department of Medicine, Division of Gastroenterology and Hepatology
Stanford University School of Medicine, Stanford University
CCSR 3115A, 269 Campus Drive
Stanford, CA 94305-5171
E-mail: jeffrey.glenn@stanford.edu
Tel: 11-650-725-3373
CHEN, OIDOVSAMBUU, ET AL. HEPATOLOGY, December 2017
1740
Materials and Methods
PATIENT SERUM SAMPLES
Deidentified unique serum samples from the follow-
ing collections were used for this study: 82 historical
HDV RNA
1
patient sera, 30 samples from HBV-
monoinfected patients, 30 samples from hepatitis C
virus (HCV)–monoinfected patients, 10 samples from
pregnant women and 10 samples from healthy control
patients, and 123 HBsAg
1
samples from a national
survey study recently conducted in Mongolia. Briefly,
for the latter, study subjects were chosen based on a
three-stage cluster sampling method to reflect the gen-
der, age, and geographical origin of the Mongolian
population. Participants were randomly selected from
the adult general population in 16 different locations,
representative of the country.
(13)
A total of 1,158 subjects were enrolled in the study,
and 123 of them tested positive for HBsAg, using a
commercial ELISA kit (DiaSorin, srl., Saluggia, Italy).
PREPARATION OF INTERNAL
STANDARD ANTI-HDV
IMMUNOGLOBULIN G
REFERENCE ANTIBODY
An internal standard of purified anti-HDV immu-
noglobulin G (IgG) was prepared from an HDV
1
sample with high-titer anti-HDV using a protein G
column. The concentration of purified IgG was deter-
mined using an Easy-Titer IgG Assay kit according to
the manufacturer’s instructions.
EXPRESSION AND PURIFICATION
OF RECOMBINANT FULL-
LENGTH S-HDAg
S-HDAg was expressed in and purified from Escher-
ichia coli and stored in single-use aliquots of >90%
purity until use, as described in the Supporting
Information.
ANTIGEN MICROARRAY
PRINTING
pGOLD slides were purchased from Nirmidas Bio-
tech, Inc. (Palo Alto, CA), containing a functionalized
coating of polyethylene glycol and terminal activated
carboxylic acid groups for amine coupling of proteins.
pGOLD slides were loaded into a microarray printing
robot (Bio-Rad) where S-HDAg (100 lM) was
printed using solid pins (Arrayit) at 25"C and 60%
humidity, resulting in microarray feature diameters of
!2 mm. The microarray layout was designed using the
microarray printer software. The antigen was printed
into 16 areas with six replicate spots each (Fig. 1A).
#######################################################################################################################################
FIG. 1. Schematic of anti-HDV Q-MAC assay and lower limit of detection using delta antigen printed on pGOLD microarray
slide. (A) S-HDAg was printed on a pGOLD microarray slide by a microarray printing robot such that six identical spots of !2 mm
diameter are contained within each partitioned area of the slide. (B) Microarray imaging results of serial dilutions of purified anti-
HDV IgG incubated on a pGOLD microarray slide, as detected by fluorescence intensity. Blank 5FBS. (C) Calibration curve for
captured anti-HDV quantification. Mean fluorescence intensity of IRDye800-labeled antihuman secondary antibody emission from
the six replicate microarray spots at each antibody concentration on the pGOLD microarray slides. Error bars represent the standard
deviation of the mean over the six replicate assay features within each partition of the slide.
#######################################################################################################################################
HEPATOLOGY, Vol. 66, No. 6, 2017 CHEN, OIDOVSAMBUU, ET AL.
1741
The slides were dried in a desiccator, vacuum-sealed in
a bag, and stored at 4"C.
Q-MAC ASSAY
The microarray printed slides were blocked with
fetal bovine serum (FBS) for 1 hour, followed by wash-
ing three times with phosphate-buffered saline con-
taining 0.5% Tween-20. Up to 13 individual serum
samples were analyzed per slide. One microliter of
each human serum sample was diluted to a total of 50
lL with FBS and applied to one well of the array for 1
hour. On each slide, the following controls were each
applied to separate wells: blank control (FBS), negative
control (HCV patient sera), and internal standard pos-
itive control (purified IgG antibody from HDV patient
sera). Slides were washed three times with phosphate-
buffered saline containing 0.5% Tween-20 and
IRDye800-labeled donkey antihuman IgG (Rockland
Immunochemicals, Inc.) diluted 1:1,000 in FBS solu-
tion and applied to each array set for 1 hour in the
dark. Slides were then washed three times with
phosphate-buffered saline containing 0.5% Tween-20
and once with deionized water and dried in the dark.
Slides were scanned using a Licor Odyssey scanner
with the 800-nm channel. Image Studio Lite, version
4.0, was used to automatically identify features above a
composite pixel intensity of 5. A fluorescence intensity
of 100 ng/mL internal standard purified anti-HDV
IgG was defined as 1 U and used to normalize the
intensity of tested samples with the following formula:
Value of fluorescence intensity (unit) 5(mean sample
exact intensity value – mean blank intensity value)/
(internal standard intensity value – mean blank intensi-
ty value).
ANTI-HDV IgG ELISA
DETECTION
Anti-HDV IgG testing was performed using a com-
mercial HDV IgG ELISA kit (GenWay Biotech,
Inc.) according to the manufacturer’s instructions. Fif-
ty microliters of patient serum was diluted 1:1 for the
ELISAs. Optical density 450 values were ascertained
using photometry (Tecan Group Ltd., Switzerland).
WESTERN BLOT DETECTION
Purified recombinant protein (S-HDAg) was sub-
jected to 12% sodium dodecyl sulfate polyacrylamide
gel electrophoresis, transferred onto a polyvinylidene
fluoride membrane (Millipore), probed with 10 lL of
patient serum diluted 1:100, followed by detection
with IRDye800-labeled goat antihuman IgG (diluted
1:20,000) and visualization using a Licor Odyssey
scanner, as described in the Supporting Information.
HDV RNA EXTRACTION AND
FULL-LENGTH GENOMIC HDV
RNA CALIBRATION STANDARD
PREPARATION
HDV RNA was extracted from 140 lL of plasma
using a QIAamp Viral RNA Mini Kit (Qiagen, Hilden,
Germany). In vitro transcription of full-length HDV
RNA from plasmid pT7GM that contains 1,679 bp of
the HDV genome was used as HDV RNA reference
standard for the quantification of HDV RNA by real-
time PCR,
(14)
as described in the Supporting
Information.
QUANTITATION OF HDV RNA BY
TAQMAN-BASED ONE-STEP
REAL-TIME PCR
In order to enable detection of all eight known gen-
otypes, the highly conserved ribozyme region of the
HDV genome was selected for design of the primers
and probes
(15)
and subsequent PCR assays, as
described in the Supporting Information. Standards
for the calibration curve were prepared using a 10-fold
dilution series of full-length HDV RNA to cover the
range 1.6 310
1
to 1.6 310
7
IU/mL. A normal
human serum negative control and an HDV
1
serum
positive control (1 310
4
RNA IU/mL) were included
in each assay. The World Health Organization HDV
RNA international standard was used to normalize the
results to international units of HDV RNA.
The linearity of the PCR ranged from 1.6 310
1
to
1.6 310
7
IU/mL.
STATISTICAL ANALYSES
Statistical analyses were performed using the Stu-
dent ttest. Receiver operating characteristic curve anal-
ysis was used to assess assay sensitivity/specificity.
Linear regression analysis was used to evaluate the cor-
relation between fluorescence intensity and HDV
RNA level. All data are reported as means 6standard
deviation. P<0.05 was considered significant.
CHEN, OIDOVSAMBUU, ET AL. HEPATOLOGY, December 2017
1742
Results
DETECTION OF ANTI-HDV
REFERENCE SERUM BY
STANDARD WESTERN BLOT
Full-length S-HDAg was expressed in BL
21(DE3) bacteria cells and purified to yield a source
of recombinant delta antigen for use in anti-HDV
detection (Supporting Fig. S1A). The recombinant
antigen was subjected to western blot analysis and
probed with serial dilutions of an anti-HDV purified
IgG internal reference standard in order to determine
the limit of detection (Supporting Fig. S1B). The
lowest detected concentration was 10 ng/mL purified
anti-HDV IgG.
DYNAMIC RANGE,
REPRODUCIBILITY, SENSITIVITY,
AND SPECIFICITY OF THE ANTI-
HDV Q-MAC ASSAY
Recombinant S-HDAg protein was printed on
pGOLD microarray slides by a microarray printing
robot such that six replicate spots were printed per
future assay area (see Fig. 1A). Following placement of
the slide’s partitions so as to create individual assay
wells, serial dilutions of anti-HDV reference IgG were
added to separate wells for 1 hour. Wells were then
washed, bound anti-HDV was detected with
IRDye800-labeled antihuman IgG, and the fluores-
cence intensity associated with each spot of printed S-
HDAg was measured (Fig. 1B). The linear range of
fluorescence intensity detection extended down to 1
ng/mL anti-HDV IgG, and the lower limit of detec-
tion was 10 pg/mL for this Q-MAC assay (Fig. 1C).
A fluorescence intensity of 100 ng/mL internal stan-
dard purified anti-HDV IgG was defined as 1 U and
used to normalize fluorescence intensity determina-
tions as detailed in Materials and Methods.
To determine the reproducibility of the new assay,
the normalized fluorescence intensities of sera from 5
negative controls (healthy humans) and 5 HDV
RNA
1
patients were screened on microarray slides in
six replicate spots in three independent experiments
(Table 1). The mean intensity of each HDV RNA
1
sample ranged from 1.742 to 8.135 U, and the coeffi-
cient of variation ranged from 6.7% to 11.1%. The
mean intensity of each negative control ranged from
0.011 to 0.029 U, and the coefficient of variation
ranged from 9.1% to 28.6%.
TABLE 1. Reproducibility of the Q-MAC Assay (Mean Microarray Fluorescence Intensity Units)
Patients
1st
Experiment
2nd
Experiment
3rd
Experiment Mean
Standard
Deviation
Coefficient
of
Variation
Positive 1 5.369
(9.7)
4.712
(8.5)
4.314
(7.7)
4.798 0.533 11.1%
Positive 2 6.227
(4.2)
7.450
(7.2)
6.260
(7.3)
7.186 0.780 10.9%
Positive 3 8.199
(6.2)
8.704
(11.2)
7.502
(12.4)
8.135 0.603 7.4%
Positive 4 1.646
(4.1)
1.871
(8.8)
1.709
(7.6)
1.742 0.116 6.7%
Positive 5 7.751
(9.1)
7.518
(2.8)
6.641
(2.3)
7.303 0.585 8.0%
Negative 1 0.009
(8.5)
0.011
(9.4)
0.012
(6.3)
0.011 0.001 9.1%
Negative 2 0.009
(9.4)
0.011
(11.3)
0.012
(8.8)
0.011 0.001 9.1%
Negative 3 0.029
(6.9)
0.032
(11.6)
0.027
(10.8)
0.029 0.003 10.3%
Negative 4 0.034
(8.8)
0.033
(2.6)
0.021
(7.1)
0.029 0.007 24.1%
Negative 5 0.027
(3.4)
0.021
(10.5)
0.015
(7.7)
0.021 0.006 28.6%
Q-MAC assay fluorescence intensity units determined in three independent assays performed on separate days on five HDV
1
samples
(Positive 1-5) and five HDV
–
samples (Negative 1-5) are displayed, along with the corresponding mean and standard deviation values.
Each value in the first three columns represents the mean fluorescence intensity of six replicate spots, with the intra-run variability (%
coefficient of variation) between each of the six spots indicated in parentheses (as indicated in Fig. 1).
HEPATOLOGY, Vol. 66, No. 6, 2017 CHEN, OIDOVSAMBUU, ET AL.
1743
The performance characteristics of the microarray
assay were determined using sera from 80 negative
controls (30 HBV monoinfected, 30 HCV monoin-
fected, and 10 each from pregnant women and healthy
human subjects) and 82 historical HDV RNA
1
patients on pGOLD microarray slides printed with S-
HDAg. All negative control intensities were below
0.090 U, and this was set as the microarray cutoff value
(Fig. 2A). The intensities of HDV RNA
1
samples
were all above this cutoff value (Fig. 2A). The mean
intensity was 6.221 U. The median intensity was 5.683
U, with a range from 0.166 to 16.91 U.
To correlate the performance of Q-MAC at detect-
ing anti-HDVs with western blot assay results, all 162
samples were subjected to standard western blot analy-
sis. Anti-HDV in the sera of the 82 HDV RNA
1
patients was detectable on westerns, while no such sig-
nal appeared for the 80 negative controls. For this sam-
ple set, a fluorescence intensity above 0.164 units
correlated with a positive western blot assay (Fig. 2B).
This intensity value could be set as the western blot
cutoff value.
COMPARISON OF Q-MAC
SENSITIVITY TO STANDARD
ELISA AND WESTERN BLOT
ASSAYS
The high sensitivity, broad dynamic range, and easy
adaptability of pGOLD microarray slides provide a
new assay format with which to screen anti-HDV IgG
accurately. To compare Q-MAC’s sensitivity to other
assays for detecting anti-HDV, we tested aliquots of
serially diluted anti-HDV reference IgG in Q-MAC,
a commercial ELISA, and western blot assays. As can
be seen from their respective limits of detection, the
Q-MAC assay’s sensitivity was 10
6
-fold and 10
3
-fold
#######################################################################################################################################
FIG. 2. Anti-HDV IgG detection in the
sera of 162 HDV RNA
1
and negative
control patients using Q-MAC and west-
ern blot assays. (A) Normalized fluores-
cence intensities of HDV
–
samples
(normal, HCV, and HBV controls) are
below the Q-MAC cutoff value (0.090
U). All of HDV RNA
1
samples are
above this cutoff value. (B) Sensitivity and
specificity of antibody detection by Q-
MAC were confirmed by western blot.
The fluorescence intensity of 0.164 corre-
sponds to a cutoff at and above which all
samples are positive by western blot.
#######################################################################################################################################
CHEN, OIDOVSAMBUU, ET AL. HEPATOLOGY, December 2017
1744
higher than the commercial anti-HDV IgG ELISA
kit and western blot analysis, respectively (Table 2).
ANALYSIS OF HDV PREVALENCE
IN MONGOLIA USING THE Q-
MAC ASSAY
We next used the HDV antibody capture assay to
determine the prevalence of HDV infection among the
123 samples identified to be HBsAg
1
in a recently col-
lected cohort of 1,158 samples from a national surveil-
lance study conducted in Mongolia to determine
prevalence rates of hepatitis virus infections.
(13)
HBsAg
1
samples were also analyzed independently
by anti-HDV western blot, a commercial anti-HDV
ELISA kit (DiaSorin), and quantitative RT-PCR
HDV RNA assays (see Table 3; Supporting Table S1).
Thirty-nine samples were below the 0.09 U Q-MAC
cutoff. All of these were negative by both western blot
and HDV RNA assays; 21% (8/39) of these samples
tested by ELISA were positive, which were interpreted
as false positives. 61% (75/123) of the samples were
above the previously defined threshold in the Q-MAC
assay for predicting western blot positivity (0.164 U)
(Fig. 3A). All of these samples were confirmed to be
positive on western blot. Thus, using the Q-MAC
threshold of 0.164 U had 100% sensitivity and 100%
specificity for predicting positivity on anti-HDV west-
ern blot. Although originally designed to accurately
substitute for the more laborious western blot assay,
the Q-MAC assay also performed quite well for pre-
dicting HDV RNA positivity.
Indeed, most of the samples with Q-MAC values of
0.164 U and above—93% (70/75)—were also found to
be HDV RNA
1
, representing 57% (70/123) of the
HBsAg
1
subjects. While all of the RNA
1
samples in
this cohort were scored as HDV
1
by Q-MAC assay,
7.1% (5/70) of the RNA
1
subjects had a false-negative
result by ELISA (Supporting Table S1).
Receiver operating characteristic analysis identified a
Q-MAC threshold of 1.659 U as having 100%
sensitivity for predicting HDV RNA positivity, with a
specificity of 94.3% (area under the curve 50.9978, P
<0.0001) (Fig. 3B). Sixty-six subjects had Q-MAC
values above 1.659, and all were confirmed to be HDV
RNA
1
. Four subjects with fluorescence intensities
above 0.164 but slightly below 1.659 were also found
to be HDV RNA
1
. The overall distribution of Q-
MAC values is graphically presented in Fig. 3C and
tabulated in Supporting Table S1.
Of note, 5 of the HDV RNA
–
patients met the Q-
MAC threshold for western blot positivity (fluores-
cence intensity $0.164 U), and all were confirmed to
be positive by western blot. Finally, nine samples were
above the 0.09 U Q-MAC cutoff value but below the
0.164 U threshold for western blot positivity.
Discussion
We describe here a new methodology, Q-MAC, for
detecting infection with HDV that is sensitive, is rap-
id, requires very small volumes of serum, and is high-
throughput in nature. Its quantitative nature and
empiric relationship to the results of standard western
blot and HDV RNA analyses enable prediction of
clinically meaningful virologic status. The results
allowed us to define a quantitative threshold of cap-
tured anti-HDV above which 100% of the samples are
positive for HDV RNA. Together, these attributes
make it ideal for analyzing patient cohorts. Indeed, we
have used this assay to determine the prevalence of
HDV among HBV-infected subjects in the largest and
most comprehensive sampling to date of the Mongo-
lian population. Most striking was the finding that
!60% of Mongolian patients with chronic hepatitis B
have evidence of HDV coinfection that is provocatively
related to the extremely high rate of hepatocellular car-
cinoma in Mongolia.
(16)
Peptide antigens have been recently described for
detection of anti-HDV.
(17)
These were limited, how-
ever, in their ability to detect all of the various
TABLE 2. Comparisons of the Lower Limit of Detection Associated With Different Methods Used to
Detect Purified Anti-HDV IgG
Positive Control Concentration
Detection
Method 100 lg/mL 10 lg/mL 1 lg/mL 100 ng/mL 10 ng/mL 1 ng/mL 10 pg/mL 1 pg/mL
Q-MAC 1111111–
Western blot 11111–––
ELISA 11–– ––––
1and – indicate a value above or below, respectively, each assay’s negative control cutoff.
HEPATOLOGY, Vol. 66, No. 6, 2017 CHEN, OIDOVSAMBUU, ET AL.
1745
genotypes that have been described for HDV.
(18)
Use
of full-length recombinant HDAg, as described here,
appears to not suffer from this limitation. Instead,
using full-length recombinant HDAg provides a
genotype-independent assay. Indeed, we have success-
fully used this assay in populations with diverse HDV
genotypes. Moreover, this assay has allowed the detec-
tion of samples that had falsely been deemed negative
by a genotype 1–specific HDV RNA assay (manuscript
in preparation).
Peptide microarrays on pGOLD substrate coating
on glass afford hundreds of fold near-infrared fluores-
cence enhancement when compared with commercial
streptavidin-coated glass.
(12)
Our Q-MAC assay is
based on the above pGOLD platform and exhibits a
detection capability down to 10 pg/mL anti-HDV
IgG concentrations. Moreover, the Q-MAC assay’s
sensitivity was 10
6
-fold and 10
3
-fold higher than a
commercial anti-HDV IgG ELISA kit and western
blot analysis, respectively.
Comparing the results of the Q-MAC assay to stan-
dard HDV western blot and quantitative RNA assays
enabled the assignment of very practical empiric quan-
titative thresholds when using the Q-MAC assay that
can be quite useful when analyzing new patient
cohorts, such as the one described here from Mongo-
lia. For example, the 0.164 U threshold for predicting
positivity on western blot was confirmed to be accurate
in this cohort. Indeed, 75/75 (100%) patients with
fluorescence intensity determinations at or above this
value were positive on western blot, and 48/48 (100%)
patients with fluorescence intensity determinations
below this value were negative on western blot. More-
over, most, but not all, of the patients with fluores-
cence intensity above 0.164 U were HDV RNA
1
.
One patient in the Mongolian cohort had an inten-
sity of 0.163. This sample was negative by both
western blot (and HDV RNA) assays, suggesting that
indeed the 0.164 intensity is empirically close to a cut-
off value for predicting positivity on standard western
blots. Of note, we have recently screened two large
cohorts from the United States and Africa containing
over 500 samples combined—all also analyzed by stan-
dard western blot—and this cutoff continues to indi-
cate the threshold for predicting positivity on western
blot (manuscripts in preparation).
This Mongolian cohort of HBsAg
1
patients was also
screened by a commercial anti-HDV ELISA kit. Some-
what alarmingly, 7.1% of patients screened by ELISA
were false negative using the HDV RNA assay as the
gold standard. All of these patients were predicted to be
RNA
1
based on the results of the Q-MAC assay.
Interestingly, one can define a fluorescence intensity
cutoff—1.659 U—that is predictive for 100% of
patients being HDV RNA
1
. Thus, the simple, rela-
tively high-throughput Q-MAC assay could prove
useful for prospectively identifying patients who have
active HDV replication or who could benefit from sub-
sequent reflex testing for HDV RNA.
Although there is a correlation between fluorescence
intensity of the Q-MAC assay and HDV RNA level
(Supporting Fig. S2), determination of HDV RNA
remains the assay of choice for monitoring response to
therapy.
Five patients in the Mongolian cohort were HDV
RNA
–
but clearly antibody-positive, as indicated by
both Q-MAC and standard western blot assays. Such
RNA-negative/antibody-positive patients have been
described before and may represent false-negative
RNA determination (due to RNA degradation during
storage, assay inaccuracy) or patients who have lost
active RNA replication but have residual antibody
levels.
Finally, due to the increased sensitivity of the Q-
MAC, there is another category of patients whose
fluorescence intensity unit values are above the negative
control threshold (0.090 U) but below the cutoff asso-
ciated with western blot positivity. The clinical signifi-
cance of these low but detectable levels of anti-HDV
antibodies, which might reflect distant infection, is at
present uncertain.
In addition to its small sample volume requirement,
ease of use, and relatively high-throughput nature, the
Q-MAC assay offers several convenient and practical
quantitative readouts. If a patient is below the Q-
MAC assay 0.09 U cutoff, one can say with certainty
that the patient has no evidence of (current or past)
HDV infection. If the patient is at or above the 0.164
TABLE 3. Summary of HDV Markers Among the
Mongolian Cohort of 123 HBsAg
1
Patients
Number Percent
Samples positive for anti-HDV
(above Q-MAC western blot
positivity threshold of 0.164 U)
75 61%
Samples positive for HDV RNA
(by quantitative RT-PCR)
70 57%
Of the total 1,158 Mongolian cohort samples, 123 were positive
for HBsAg. The HBsAg
1
samples were tested for anti-HDV
and HDV RNA by Q-MAC and quantitative RT-PCR assays,
respectively. The total number and percent of the 123 samples
that were positive for anti-HDV antibody and HDV RNA are
indicated. See Supporting Table S1 and text for additional
details.
CHEN, OIDOVSAMBUU, ET AL. HEPATOLOGY, December 2017
1746
U western cutoff, the patient definitely has been
infected with HDV. If above the RNA cutoff of 1.659
U, there is a 100% chance of being RNA
1
.
Various prior studies in selected Mongolian popula-
tions have indicated a range of HDV prevalences, with
the latter being consistently higher than is typical of
Western populations, although this may be limited by
suboptimal assays or sampling bias. For example, a
study on 249 apparently healthy individuals in and
around Ulaanbaatar (age 48.4 613.9 years) detected
#######################################################################################################################################
FIG. 3. Analysis of anti-HDV IgG and
HDV RNA in the sera of 123 HBsAg
1
samples from Mongolia by Q-MAC,
western blot, and qRT-PCR assays. (A)
Q-MAC assay of samples categorized by
western blot status. Note all samples
below the Q-MAC 0.164 U predictive
cutoff value for western blot positivity
were negative on western blot, and all
samples above this cutoff were positive on
western blot. (B) Receiver operating char-
acteristic analysis of Q-MAC assay fluo-
rescence intensity for predicting HDV
RNA positivity in sera of HBsAg
1
patients from Mongolia. The fluorescence
intensity value of 1.659 U was identified
as the optimal cutoff value. (C) Q-MAC
assay of samples categorized by HDV
RNA status. Fluorescence intensity cutoff
values above which 100% of samples are
predicted to be positive on western blot
or quantitative RT-PCR for HDV RNA,
respectively, are indicated, along with the
Q-MAC assay’s negative control cutoff.
Abbreviation: AUC, area under the curve.
#######################################################################################################################################
HEPATOLOGY, Vol. 66, No. 6, 2017 CHEN, OIDOVSAMBUU, ET AL.
1747
HDV RNA in 8.0% (20/249) of the participants and
in 83% (20/24) of the HBsAg
1
subjects,
(6)
although
the details of how these subjects were chosen are not
clear. Among 289 first-time blood donors at a single
center in Ulaanbaatar (age 28.9 69.6 years), HDV
RNA was detected in 26 (9%) of the total donors and
in 87% of the HBsAg
1
subjects.
(19)
In a study of 207
patients with known liver disease, 144 were HBsAg
1
,
including 117 (81% of those HBsAg
1
) with detectable
HDV RNA.
(20)
Among 655 apparently healthy chil-
dren (0.3-15 years), 64 (9.8%) were infected with
HBV and, of these, 13 (20.3%) were HDV RNA
1
.
(21)
Focusing on apparently healthy children 7-12 years old
born after 1991 (when universal vaccination against
HBV was introduced into the country), 59 of 1,182
(5%) children were found to be HBsAg
1
, and
8 (13.6%) of these were HDV RNA
1
.
(22)
An apparent
beneficial effect of vaccination is encouraging, yet its
successful implementation appears to be incomplete.
Equally concerning are the above studies pointing to a
very high prevalence of chronic HDV infection among
the HBsAg
1
adult population and the question of
whether it extends beyond the capital city’s environs.
The present study sought to address this by deter-
mining the prevalence of HDV infection within the
largest cohort to date of a prospectively randomly sam-
pled population throughout Mongolia. As such, the
results likely represent the most accurate representation
of the true prevalence rates for the assessed important
human viral pathogens. While the full description of
the HCV and HBV prevalence rates in this cohort is
described elsewhere,
(13)
we report here on the preva-
lence of HDV in Mongolia, which is astonishing.
Indeed, while the average global HDV coinfection rate
among HBV-infected subjects is estimated at 5%,
(23)
approximately 60% of HBV-infected patients are coin-
fected with HDV. Extrapolation to the general popu-
lation results in an estimated prevalence among all
Mongolian adults of 6.4 60.7% anti-HDV positivity
and 6.1 60.7% with detectable HDV RNA (Support-
ing Table S2). Reasons for this much higher preva-
lence may include the relative isolation of this
population combined with inadequate control of hori-
zontal transmission associated with dental/medical
procedures and sexual activity.
(13,22)
In any case, these
results have important implications for both public
health and the agencies and institutions concerned
with crafting and implementing appropriate responses.
In summary, we developed a quantitative microarray
anti-HDV capture (Q-MAC) assay. Defining in the
new assay quantitative thresholds of captured anti-
HDV above which 100% of the samples are positive
for anti-HDV or HDV RNA allowed for prospective
prediction of both western blot positivity and HDV
RNA positivity, respectively. This assay allowed con-
firmation of a strikingly high !60% prevalence of
HDV coinfection among HBsAg
1
subjects and a gen-
eral HDV prevalence of 6.4% among all adults in
Mongolia. The extremely high prevalence rate for this
hepatitis virus associated with increased cancer risk
(24)
may contribute to the alarming incidence of hepatocel-
lular carcinoma in Mongolia, which ranks among the
highest in the world.
(25)
This serves to underscore the
urgent need for improved therapies for HDV.
REFERENCES
1) Lin HH, Lee SS, Yu ML, Chang TT, Su CW, Hu BS, et al.
Changing hepatitis D virus epidemiology in a hepatitis B virus
endemic area with a national vaccination program. HEPATOLOGY
2015;61:1870-1879.
2) Romeo R, Del Ninno E, Rumi M, Russo A, Sangiovanni A, de
Franchis R, et al. A 28-year study of the course of hepatitis delta
infection: a risk factor for cirrhosis and hepatocellular carcinoma.
Gastroenterology 2009;136:1629-1638.
3) Heller T, Koh C, Glenn JS. Hepatitis D. In: Sanyal AJ, Boyer
T, Terrault N, Lindo K, eds. Zakim and Boyer’s Hepatology,
7th ed. New York: Elsevier; 2017;34:1-15.
4) Rizzetto M. Hepatitis D: thirty years after. J Hepatol 2009;50:
1043-1050.
5) Hughes SA, Wedemeyer H, Harrison PM. Hepatitis delta virus.
Lancet 2011;378:73-85.
6) Takahashi M, Nishizawa T, Gotanda Y, Tsuda F, Komatsu F,
Kawabata T, et al. High prevalence of antibodies to hepatitis A
and E viruses and viremia of hepatitis B, C, and D viruses
among apparently healthy populations in Mongolia. Clin Diagn
Lab Immunol 2004;11:392-398.
7) Hoofnagle JH. Type D (delta) hepatitis. JAMA 1989;261:1321-
1325.
8) Olivero A, Smedile A. Hepatitis delta virus diagnosis. Semin
Liver Dis 2012;32:220-227.
9) Le Gal F, Brichler S, Sahli R, Chevret S, Gordien E. First inter-
national external quality assessment for hepatitis delta virus RNA
quantification in plasma. HEPATOLOGY 2016;64:1483-1494.
10) Andernach IE, Leiss LV, Tarnagda ZS, Tahita MC, Otegbayo
JA, Forbi JC, et al. Characterization of hepatitis delta virus in
sub-Saharan Africa. J Clin Microbiol 2014;52:1629-1636.
11) Tabakman SM, Lau L, Robinson JT, Price J, Sherlock SP,
Wang H, et al. Plasmonic substrates for multiplexed protein
microarrays with femtomolar sensitivity and broad dynamic
range. Nat Commun 2011;2:466.
12) Zhang B, Jarrell JA, Price JV, Tabakman SM, Li Y, Gong M,
et al. An integrated peptide-antigen microarray on plasmonic
gold films for sensitive human antibody profiling. PloS One
2013; 8:e71043.
13) Dashtseren B, Bungert A, Dashdorj N. Confirmed high preva-
lence of hepatitis B and C in Mongolia: a nationwide survey
among 1158 Mongolian adults. J Med Virol. Submitted for
publication.
CHEN, OIDOVSAMBUU, ET AL. HEPATOLOGY, December 2017
1748
14) Glenn JS, Taylor JM, White JM. In vitro–synthesized hepatitis
delta virus RNA initiates genome replication in cultured cells.
J Virol 1990;64:3104-3107.
15) Karatayli E, Altunoglu YC, Karatayli SC, Alagoz SG, Cinar K,
Yalcin K, et al. A one step real time PCR method for the quan-
tification of hepatitis delta virus RNA using an external armored
RNA standard and intrinsic internal control. J Clin Virol 2014;
60:11-15.
16) Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers
C, et al. GLOBOCAN 2012: Estimated Cancer Incidence,
Mortality, and Prevalence Worldwide in 2012 [computer
program]. Version 1.0. Available at: http://globocan.iarc.fr;
2013.
17) Alves C, Cheng H, Roder H, Taylor J. Intrinsic disorder and
oligomerization of the hepatitis delta virus antigen. Virology
2010;407:333-340.
18) Radjef N, Gordien E, Ivaniushina V, Gault E, Ana€ıs P, Drugan
T, et al. Molecular phylogenetic analyses indicate a wide and
ancient radiation of African hepatitis delta virus, suggesting a
deltavirus genus of at least seven major clades. J Virol 2004;78:
2537-2544.
19) Tsatsralt-Od B, Takahashi M, Nishizawa T, Inoue J, Ulaankhuu
D, Okamoto H. High prevalence of hepatitis B, C and delta
virus infections among blood donors in Mongolia. Arch Virol
2005;150:2513-2528.
20) Tsatsralt-Od B, Takahashi M, Nishizawa T, Endo K, Inoue J,
Okamoto H. High prevalence of dual or triple infection of
hepatitis B, C, and delta viruses among patients with chronic liv-
er disease in Mongolia. J Med Virol 2005;77:491-499.
21) Tsatsralt-Od B, Takahashi M, Endo K, Agiimaa D, Buyankhuu
O, Ninomiya M, et al. Prevalence of hepatitis B, C, and delta
virus infections among children in Mongolia: progress in child-
hood immunization. J Med Virol 2007;79:1064-1074.
22) Davaalkham D, Ojima T, Uehara R, Watanabe M, Oki I,
Nymadawa P, et al. Hepatitis delta virus infection in Mongolia:
analyses of geographic distribution, risk factors, and disease sever-
ity. Am J Trop Med Hyg 2006;75:365-369.
23) Wedemeyer H, Manns MP. Epidemiology, pathogenesis and
management of hepatitis D: update and challenges ahead. Nat
Rev Gastroenterol Hepatol 2010;7:31-40.
24) Kushner T, Serper M, Kaplan DE. Delta hepatitis within the
Veterans Affairs medical system in the United States: prevalence,
risk factors, and outcomes. J Hepatol 2015;63:586-592.
25) Alcorn T. Mongolia’s struggle with liver cancer. Lancet 2011;
377:1139-1140.
Author names in bold designate shared co-first
authorship.
Supporting Information
Additional Supporting Information may be found at
onlinelibrary.wiley.com/doi/10.1002/hep.28957/suppinfo.
HEPATOLOGY, Vol. 66, No. 6, 2017 CHEN, OIDOVSAMBUU, ET AL.
1749
